The proposed research seeks to: evaluate and compare the effectiveness of two common pharmacological treatments (Imipramine and Oxybutynin) with a non pharmacological treatment (Full-Spectrum Home Training) for childhood primary enuresis compare side effects and compliance for the two medications; assess effects of pharmacological treatment failure on subsequent behavioral treatment; and provide two year follow up evaluation of the psychosocial adjustment correlates of treatment success and failure. Large scale controlled outcome trials comparing Imipramine with Oxybutynin have not been conducted despite the fact that increasing numbers of enuretic children are receiving Oxybutynin as an alternative to medical management with Imipramine. Further recent studies have raised questions about the widespread use of Imipramine for enuresis, particularly concerning possible cardiovascular side effects. It is unknown whether these are durable or transient effects. Since it is likely that most enuretic children will continue to receive medication treatment prior to behavioral treatment and because previous research suggests some negative effects of Imipramine treatment failure on subsequent behavioral treatment, it is important to experimentally evaluate the effects of medication failure on behavioral treatment. In order to make the-public and professionals aware of the risks and benefits of actively recommending treatment for bedwetting, it is important to document the effects of treatment success and treatment failure on children's psychosocial functioning beyond mere post treatment follow up. The research design is a 3 group randomized experiment. 225 enuretic children will be randomly assigned to Imipramine, Oxybutynin, or Full-Spectrum treatment during Phase I. Psychological and physiological assessments will be conducted throughout. All children failing to remit from pharmacological treatments and those who relapse from medication will receive behavioral treatment. We will assess physiological changes in bladder functioning and psychosocial functioning at 0,3,6,12,18, and 24 months. Power calculations show that we can detect projected differences in categorical determinations of outcome and even smaller effects for continuous variables. Prediction equations will be developed to guide professionals in identifying children who are most likely to benefit from the treatment approaches.